Banca de DEFESA: Vitor Pereira Silva

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : Vitor Pereira Silva
DATE: 31/07/2023
TIME: 14:00
LOCAL: PECC
TITLE:

PREDICTION OF CONVENTIONAL CONCRETE PROPERTIES THROUGH THE USE OF MACHINE LEARNING TECHNIQUES

 

KEY WORDS:

Machine learning; cementitious materials; concrete; properties; compressive strength.

PAGES: 92
BIG AREA: Engenharias
AREA: Engenharia Civil
SUBÁREA: Construção Civil
SPECIALTY: Materiais e Componentes de Construção
SUMMARY:

Recently, several machine learning (ML) techniques are emerging as alternative and efficient ways to predict how component properties influence the properties of the final mixture. In the area of civil engineering, recent research already uses ML techniques in relation to conventional concrete dosages. The importance of discussing its use in the Brazilian context is inserted in an international context in which this methodology is already being applied, and it is necessary to verify the applicability of these techniques with national databases or with what is created from national input data. In this research, one of these techniques, an artificial neural network (ANN), is used to determine the compressive strength of conventional Brazilian concrete at 7 and 28 days, using a database built through publications in congresses and academic papers and comparing -o with Yeh's reference database. The data were organized into nine variables and five different cases where the data samples used for training and testing vary. The eight possible input variables were: cement consumption, blast furnace slag, pozzolana, water, additive, fine aggregate, coarse aggregate and age. The response variable was the compressive strength of the concrete. Using international data as a training set and Brazilian data as a test set, or vice versa, did not show satisfactory results in isolation. The results showed a variation in the five scenarios; however, when using the Brazilian and reference database together as test and training sets, an R² of 0.97 and an R² of 0.86 were obtained, showing that, in the union of the two databases, a good predictive model is obtained.

COMMITTEE MEMBERS:
Presidente - 1734221 - JOAO HENRIQUE DA SILVA REGO
Interna - 1534368 - MICHELE TEREZA MARQUES CARVALHO
Externa à Instituição - JULLIANA SIMAS VASCONCELLOS - UEG